It is common for communications devices to have multiple antennas that are packaged close together (e.g., less than a quarter of a wavelength apart) and that can operate simultaneously within the same frequency band. Common examples of such communications devices include portable communications products such as cellular handsets, personal digital assistants (PDAs), and wireless networking devices or data cards for personal computers (PCs). Many system architectures (such as Multiple Input Multiple Output (MIMO)) and standard protocols for mobile wireless communications devices (such as 802.11n for wireless LAN, and 3G and 4G data communications such as 802.16e (WiMAX), HSDPA, 1xEVDO, and LTE) may require multiple antennas operating simultaneously.
The trend in mobile wireless devices has been to provide faster access, improved processors, more memory, brighter and higher resolution screens, additional connectivity with Wi-Fi, GPS, 3G and 4G world access—all with longer battery life in thinner more sleek packages. Compound this with the desire of mobile operators to expand their available band allocations, and what results is a difficult industrial design arena, where suppliers are vying for physical space within the confines of a smartphone or similar device to accommodate necessary components. One such component is the antenna—essentially a transducer that converts time varying electrical current to radiated energy, and often considered a last minute addition to the physical structure.
A number of factors can affect antenna performance in a mobile communication device. While these factors are related, they generally fall into one of three categories: antenna size, mutual coupling between multiple antennas, and device usage models.
The size of an antenna can be dependent on three criteria: bandwidth of operation, frequency of operation, and required radiation efficiency. Bandwidth requirements have increased as they are driven by FCC frequency allocations in the US and carrier roaming agreements around the world. Different regions use different frequency bands, now with over 40 E-UTRA band designations-many overlapping but requiring world capable wireless devices to typically cover a frequency range from 698 to 2700 MHz.
A simple relationship exists between the bandwidth, size, and radiation efficiency for the fundamental or lowest frequency resonance of a physically small antenna.
                                          Δ            ⁢                                                  ⁢            f                    f                ∝                                            (                              a                λ                            )                        3                    ⁢                      η                          -              1                                                          (        1        )            
The variable a is the radius of a sphere containing the antenna and its associated current distribution. Since a is normalized to the operating wavelength λ, the formula may be interpreted as “fractional bandwidth is proportional to the wavelength normalized modal volume”. The radiation efficiency η is included as a factor on the right side of equation (1), indicating that greater bandwidth is achievable by reducing the efficiency. Radio frequency currents exist not only on the antenna element but also on the attached conductive structure or “counterpoise”.
For instance, mobile phone antennas in the 700-960 MHz bands can use an entire PCB as a radiating structure so that the physical size of the antenna according to equation (1) is actually much larger than what appears to be the “antenna”. The “antenna” may be considered a resonator that is electromagnetically coupled to the PCB so that it excites currents over the entire conductive structure or chassis. Some smartphones exhibit conductive chassis dimensions of approximately 60×110 mm, which from an electromagnetic modal analysis predicts a fundamental mode somewhat over 1 GHz suggesting that performance bandwidth degrades progressively at lower excitation frequencies. The efficiency-bandwidth trade-off is complex requiring E-M simulation tools for accurate prediction. Results indicate that covering 698-960 MHz (Bands 12, 13, 17, 18, 19, 20, 5 and 8) with a completely passive antenna with desirable antenna size and geometry becomes difficult without making sacrifices in radiation efficiency.
Factors determining the achievable radiation efficiency are not entirely obvious, as the coupling coefficient between the “antenna” and the chassis, radiative coupling to lossy components on the PCB, dielectric absorption in plastic housing, coupling to co-existing antennas, as well as losses from finite resistance within the “antenna” resonator structure, all play a part. In most cases, the requirements imposed by operators suggest minimum radiation efficiencies of 40-50%, so that meeting a minimum total radiated power (TRP) requirement essentially requires tradeoffs between the power amplifier (PA) output and an achievable antenna efficiency. In sum, poor efficiency at the antenna translates to less battery life, as the PA must compensate for the loss.
Prior to requiring band aggregation, wireless devices operated on one band at a time with need to change when roaming. Consequently, the required instantaneous bandwidth would be considerably less than that required to address worldwide compatibility. Take a 3G example for instance, where operation in band 5 from (824-894 MHz) compared to operation in bands 5 plus 8 (824-960 MHz). Then, add the requirements for band 13 and band 17 and the comparison becomes more dramatic—824-960 vs. 698-960 MHz. This becomes problematic as legacy phone antennas support pentaband operation but only bands 5 and band 8.
Given equation (1) several choices exist. One choice would be to increase the antenna system size, (i.e. the antenna and phone chassis footprint) and/or to reduce the radiation efficiency. Since 4G smartphones require 2 antennas, neither approach is necessarily desirable from an industrial design standpoint, although it is possible to cover the 700-2200 MHz bands with a completely passive antenna in a space allocation of 6.5×10×60 mm.
Various alternative antenna configurations motivated by industry are the following:                Limit the antenna(s) instantaneous bandwidth within current antenna space allocations to allow use of 1 or more antennas without compromising the industrial design (Antenna Supplier motivation)        Make the antenna(s) smaller to achieve a compact and sleek device with greater functionality by limiting the instantaneous bandwidth with same or improved antenna efficiency (OEM motivation)        Improve the antenna efficiency, and therefore the network performance by controlling the antenna instantaneous frequency/tuning (Operator motivation)        Make the antenna agile to adapt to different usage models (OEM/User/Operator motivation)        Combinations of the above        
One approach can be to limit the instantaneous operation to a single band to satisfy the protocol requirements for a single region. To satisfy the roaming requirements, the antenna could be made frequency agile on a band-by-band basis. This approach represents the most basic type of “state-tuned” antenna.
Unfortunately, mobile operators are now vying for more bandwidth, and are finding it necessary to use more than one frequency band simultaneously for transmission and/or reception of data. This is generally known as inter-band carrier aggregation (CA), which can provide the operator with the option to use many simultaneous channels for high bandwidth data transmission and/or to expand lower bandwidth channels into alternative spectral regions to off load network congestion. Exactly how these channels are used is still to be determined, and many options exist. One such option is to use more than one channel in one band—intra-band carrier aggregation in conjunction with inter-band CA.
For CA to be effective, however, all frequency bands should be functional at the same time, which implies that tuning an antenna may not necessarily be desirable since tuning often implies optimizing one or more bands for performance while leaving others at a performance disadvantage. The ideal antenna would cover all bands at the same time allowing the operator to operate in any band or frequency region simultaneously.
An additional option to expand data throughput is through the use of channel independence to send multiple streams of independent data, as Multiple Input Multiple Output (MIMO) signal transmission or Single Input, Multiple Output (SIMO), or Multiple Input, Single Output (MISO) transmission. To achieve a high data rate, signals must maintain low inter-signal correlation, requiring that antennas used for reception or transmission do not mix those signals that would normally be decorrelated via random perturbations experienced over the reception or transmission path through space.
Previous solutions using multi-band antennas have not always provided adequate decorrelation when more than two antennas are implemented on the same platform. This is due to sharing of currents developed on the common ground plane printed circuit when antennas are simultaneously excited. Placement of antennas therefore becomes difficult especially when multiple antennas need to be separately located.